Maximum probabilities, information, and choice under uncertainty

Abstract

This note proposes a simple, expected utility-like model for decision making under uncertainty. The model uses the maximum probability for each possible outcome and the amount of information conveyed by this upper envelope. A graphical tool is introduced and used to study the model when two outcomes are possible. The model is extended to an abstract number of outcomes in which interpersonal comparisons of preferences are considered along with applications to medical decision making and financial asset demand.